Table 1 Studies included in the review of climate impact on dengue

Earnest et al. [32]

Singapore (2001–2008)

Weekly laboratory confirmed notified dengue cases

Weekly climate (mean/minimum/maximum temperature, mean rainfall, mean/minimum/maximum relative humidity, mean hours of sunshine and mean hours of cloud) data

Local meteorological station data

Poisson regression, Sinusoidal function

Temperature, relative humidity and SOI associated with dengue cases.

Temporal trends of dengue were noticeable.

Descloux et al. [31]

Noumea (New Caledonia) (1971–2010)

Monthly confirmed cases of DF/ DHF

Monthly climate (temperature, precipitation, relative humidity, wind force, potential evapo-transpiration, hydric balance sheet) data and ENSO indices

Local meteorological station data

Non-linear models

Significant inter-annual correlations were observed between dengue outbreaks and summertime temperature, precipitation, relative humidity but not ENSO.

The epidemic dynamics of dengue were driven by climate.

Chen et al. [30]

Taiwan (1994–2008)

Daily confirmed cases of notified DF

Daily climate (temperature, rainfall) data, socio-demographic factors

Local meteorological station data

GAM

Rainfall was correlated with dengue cases. Lag effects were observed.

A climatic change does have influence on dengue outbreaks.

Hu et al. [43]

Australia (2002–2005)

Monthly confirmed cases of notified dengue

Monthly weather, SEIFA, pop (LGA)

Local meteorological station data

Bayesian CAR

Increase in dengue cases of 6% in association with a 1-mm increase in average monthly rainfall and a 1°C increase in average monthly maximum temperature, respectively was observed.

Socio-ecological factors appear to influence dengue. The drivers may differ for local and overseas cases. Spatial clustering of dengue cases was evident.

Chowell [45]

Peru (1994–2008)

Annual confirmed cases

Time series of annual population size and density, altitude and climate data

Local meteorological station data

Wavelet time series

A significant difference in the timing of epidemics between jungle and coastal regions was observed.

The differences in the timing of dengue epidemics between jungle and coastal regions were significantly associated with the timing of the seasonal temperature cycle.

Thai et al. [50]

Vietnam (1994–2009)

Monthly confirmed cases

Monthly climate (mean temperature, rainfall and relative humidity) data and ENSO indices

Local meteorological station data

Wavelet time series

ENSO indices and climate variables were significantly associated with dengue incidence.

Climate variability and ENSO impact dengue outbreaks.

Colon-Gonzalez et al. [41]

Mexico (1985–2007)

Monthly confirmed cases

Monthly climate (minimum and maximum temperature and rainfall) and ENSO indices

Local meteorological station data

Linear regression, Phillips–Perron and Jarque–Bera test tests

Incidence was higher during El-Nino. Incidence was associated with El-Nino and temperature during cool and dry times.

Temperature was an important factor in the dengue incidence in Mexico.

Pinto et al. [9]

Singapore (2000–2007)

Weekly confirmed notified DF cases

Weekly climate (maximum and minimum temperature, maximum and minimum relative humidity) data

Local meteorological station data

Poisson regression, Principal component anlaysis

For every 2–10 degrees of maximum and minimum temperature variation, an increase of cases of 22-184% and 26-230% respectively, was observed.

Temperature was the best predictor for the dengue increase in Singapore.

Gharbi et al. [36]

French West Indies (2000–2007)

Weekly laboratory confirmed cases from hospitals or not

Weekly climate (cumulative rainfall, relative humidity, minimum, maximum and average temperature) data

Local meteorological station data

Time series (SARIMA), RMSE and Wilcoxon signed-ranks test

Temperature was significantly associated with dengue forecasting but not humidity.

Temperature improves dengue outbreaks better than humidity and rainfall.

Hu et al. [42]

Australia (1993–2005)

Monthly confirmed cases of notified DF cases

Monthly SOI, rainfall and annual population

Local meteorological station data

Cross-correlations, Time series (SARIMA)

A decrease in the SOI was significantly associated with an increase in the dengue cases.

Climate variability is directly and/or indirectly associated with dengue. SOI based epidemic forecasting is possible.

Johansson et al. [48]

Puerto Rico, Mexico, Thailand (1986–2006)

Monthly reported cases of DF/ DHF

Monthly climate (precipitation, minimum, maximum and mean average temperature) data and ENSO indices

Global climate surfaces (0.5 × 0.5°) local meteorological station data

Wavelet time series

Temperature, rainfall and dengue incidence were strongly associated in all three countries for the annual cycle. The associations with ENSO varied between countries in the multi-annual cycle.

The role of ENSO may be obscured by local climate heterogeneity, insufficient data, randomly coincident outbreaks, and other, potentially stronger, intrinsic factors regulating dengue transmission dynamics.

Bambrick et al. [44]

Australia (1991–2007)

Annual incidence – notified cases of DF

Annual Temperature, vapour pressure and population

Local meteorological station data

Climate change scenarios

Geographic regions with climates that are favourable to dengue could expand to include large population centres in a number of currently dengue-free regions.

An eight-fold increase in the number of people living in dengue prone regions in Australia will occur unless greenhouses gases are reduced.

Bulto et al. [46]

Cuba (1961–1990)

Dengue-specific parameters of DF/ DHF

Monthly climate (maximum and minimum temperature, precipitation, atmospheric pressure, vapour pressure, relative humidity, thermal oscillation and solar radiation) data

Local meteorological station data

Multivariate (Empiric orthogonal function)

Strong associations between climate anomalies and dengue

Climate variability has influence on dengue.

Cazelles et al. [40]

Thailand (1983–1997)

Monthly confirmed cases of DHF

Monthly climate (temperature and rainfall) data and ENSO indices

Local meteorological station data

Wavelet time series

Strong association between dengue incidence and El-Nino events was observed. Temperature had greater influence on dengue than rainfall.

The association is non-stationary and have a major influence on the synchrony of dengue epidemics.

Hales et al. [33]

Global (1975–1996)

Monthly reported cases of DF

Monthly climate (maximum, minimum and mean temperature, rainfall and vapour pressure) and population and projections (future climate and population) data

Region-specific and GCM projections

Logistic regression and IPCC scenarios

In 2085, under climate and population projections, 50-60% of the population would be at dengue risk.

There is a potential increase in the dengue risk areas under climate change scenarios, if the risk factors remain constant.

Patz et al. [38]

Global (1931–1980)

Dengue-specific parameters

Monthly climate data

Site-specific GCM

GCM output to vectorial capacity

Among the three GCMs, the average projected temperature elevation was 1.16°C, expected by the year 2050.

Epidemic potential increased with a relatively small temperature rise, indicating that lower mosquitoes infestation values would be necessary to maintain or spread dengue in a vulnerable population.